Need good suspension geometry simulator: advice please

[gamesjunston]

I'm doing a chassis design project for my major project in my degree (2nd year). The design uses pushrod suspension.

As well as creating a 3d cad model (Autodesk Inventor Prof 2012) of my chassis design I am looking to create a simulation of the suspension geometry and if possible calculate the geometry values.

I'm also doing an FEA simulation but I can do this with the college CAD software.

I need to know if SolidWorks has the necessary simulation capabilities – as well as if my Autodesk 3d CAD model can be easily transferred – before I go ahead and buy it.

Can anyone tell me if SolidWorks is good for these purposes and if it is likely to take a long time to learn?

Failing this can anyone recommend any other programs to simulate suspension geometry from CAD models?

Many thanks.

 

[sbozy25]

If you have your suspension designed and drawn correctly, it is only a matter of putting the right constraints on the assembly. Then set up your different conditions... Full wheel cut, full jounce, rebound, single wheel event, etc. The sky is the limit...

Cabbages, knickers, It hasn't got A BEAK!

 

[GregLocock]

Any 3D CAD package should be able to be forced to generate the kinematics of a sufficiently simple suspension geometry. They do have problems with many real world suspensions, and they don't automatically give all the measurements that suspension designers need.

Inventor certainly can be made to do animated linkages, and I don't actually see much value in transferring to SW if you don't already know it, but that's just a religious war. Whether SW can smoothly import your model depends on a lot of things, but at the very least you'll lose your history tree.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[patprimmer]

Is this a formula SAE project?

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[GregLocock]

Judging by Facebook, yes probably. In which case you should be aware that many teams (ab)use ADAMS and you have or had a perfectly good suspension/dynamics forum on the fsae site.

http://fsae.com/eve/forums/a/tpc/f/125607348/m/33320056051

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Komodo86]

I done mine in Pro/Engineer, but as a poster above said, it's only a matter of constraining the model correctly, any package should be able to do it. The limitations (as far as I can see) are that you are forced to accept results with zero compliance. This might be acceptable for a purpose built race car using spherical links everywhere, but for a road car using a lot of rubber components, it does cause some problems.

If anyone has any solutions to this then I am interested to hear them!

 

[WolfHR]

I think even compliant suspensions could be 'simulated' in CAD package like CATIA or pro/E. The problem boils down to two issues- the simpler one is question of moving pivot points. This can be done using formulas to define pivot point coordinates (e.g. P = P0 +c*F), but would to a certain extent conflict with 'automated' process of suspension modeling I've used in CATIA (well, that's not very relevant, but hey).

The more problematic issue would be necessary calculations of forces acting on bushings, but I'd think it would be solvable. I'm not sure either of the CADs is equipped with matrix operations so it would boil down to forcing them to solve 6 equations with 6 unknowns 'manually' to obtain forces in outer ball joints. Doable, but not very practical.

What I would find most problematic is that this would not be 'exact' or 'correct' method- and I don't think the most correct way could be made in CAD at modeling level. The reason for that is that one should (if forces at CP are known) and knowing the spring rate calculate the position at which the 'system' is in equilibrium (and I do not think iterative calculations are possible in either of the packages). The best one could do, as I see it, is to do iterative part manually- 'clamp' suspension at one point, let CAD calculate all the stuff and the required spring rate, and adjust bump until calculated rate is close enough to actual value. And to make sure the program's initial guess is close to the mark (relatively accurate way of modeling the wheel rate).

... hmmm, maybe I should try to put my money where my mouth is ... (all this rambling of mine jolted a few remaining brain cells and made me wonder if one could use sketcher to compute some of the forces, making things much less complex and 'math intensive', providing a more elegant model)

 

[GregLocock]

Note that even real suspension analysers often seem to ignore this.

For example I don't know how Susprog copes with over-constrained suspensions, or with those where the bush compliances are large, or where the flexibility of the arms is part of the kinematics.

It is also worth pointing out that rubber bushes have non linear characteristics.





Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Moparigno]

I also use Autocad Inventor 2012 for chassi design.

But my advice is to use a specific software for suspension design and simulation. They are more easy to use and provides better results.

Take a look to this;

http://www.optimumg.com/

 

[GregLocock]

I think you'll find OptimumK only handles a small subset of suspension designs, and does not allow for compliances. As such it doesn't seem to do anything that an experienced CAD guy couldn't do in Inventor. I may be wrong, having never used it.

The compliance issue is a really big deal in some circumstances, even for macphersons.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[gamesjunston]

Guys, sorry for not saying thanks before now, really useful advice, many thanks for responding.

As it happens my university does have 2 Formula Renault cars which it has raced in previous years, though not since I started my course have either car been used for competition / FSAE due to funding issues (very frustrating!)

The purpose is a major project for a chassis design, I'll be using Inventor Professional 2012 to model it (as part of another objective) so wanted to know what the easiest method / package to simulate kinematics / dynamics might be as I need some sort of results or data from my own chassis design to compare with real-world examples - using the Elise as my benchmark - so that it can be quantified as a successful design in some way.

I now know the basics of Inventor's dynamic simulation feature, as well as FEA which I'll also be using to hopefully determine chassis stiffness.

Sounds like if I use it properly I can get certain things from Inventor but not as much as if I used specialised software.

The trouble I've got is that I've not much time left to finish the project (around 4 weeks) but I need to able to get enough data to work with. Also not all of the packages out there will be available to me unless I can get a cracked copy, free trial or significant student discount (recommendations are welcome all the same, mind you). The parameters I was initially hoping to determine were:

- C of G longitudinal & lateral through full compression
- camber through full compression
- toe through full compression
- roll stiffness
- weight transfer under roll and pitch
- wheel motion ratio
- roll motion ratio
- chassis torsional rigidity (FEA)
- mounting point stress / stiffness / deflection (FEA)

The FEA stuff I’m confident I can use Inventor for, it’s the suspension stuff I’m not so sure about. I’d be incredibly surprised if I can simulate all of the parameters I’ve listed but if I can get most of them with Inventor I’ll definitely make do with that.

Does it sound like I'll be able to get away with Inventor or is it just too limited for what I want to find?

Many thanks again for the help.

 

[GregLocock]

In 4 weeks you don't have time to get reliable results from the more advanced packahges.

- C of G longitudinal & lateral through full compression
amazingly unimportant, handcalcs

- camber through full compression
Inventor

- toe through full compression
Inventor

- roll stiffness
Inventor to get spring lengths etc then hand calcs
- weight transfer under roll and pitch
cgz/track etc - handcalcs
- wheel motion ratio
inventor
- roll motion ratio
inventor
- chassis torsional rigidity (FEA)

- mounting point stress / stiffness / deflection (FEA)

Incidentally word to the wise, we don't do cracks on eng-tips, even if they aren't illegal many of us write software for a living.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[gamesjunston]

Thanks Greg,

sorry about that, I prefer not to use cracks myself but student loans don't go that far which means having to stoop a bit lower in desperate times. Solidworks bring the price down to around £100 for a 3 month license for students, that kind of thing I'd always fork out for (if I knew it was the right package).

Interesting to know what can be done using handcalcs, I actually borrowed A.Staniforth 'Competition Car Suspension' a while ago and I think that gave a good list of calculations, may well use that again.

Many thanks